File size: 24,227 Bytes
2c9721b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 |
2023-10-18 17:39:46,413 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,413 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 17:39:46,413 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,413 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
- NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
2023-10-18 17:39:46,413 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,413 Train: 3575 sentences
2023-10-18 17:39:46,413 (train_with_dev=False, train_with_test=False)
2023-10-18 17:39:46,413 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,413 Training Params:
2023-10-18 17:39:46,414 - learning_rate: "5e-05"
2023-10-18 17:39:46,414 - mini_batch_size: "4"
2023-10-18 17:39:46,414 - max_epochs: "10"
2023-10-18 17:39:46,414 - shuffle: "True"
2023-10-18 17:39:46,414 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,414 Plugins:
2023-10-18 17:39:46,414 - TensorboardLogger
2023-10-18 17:39:46,414 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 17:39:46,414 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,414 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 17:39:46,414 - metric: "('micro avg', 'f1-score')"
2023-10-18 17:39:46,414 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,414 Computation:
2023-10-18 17:39:46,414 - compute on device: cuda:0
2023-10-18 17:39:46,414 - embedding storage: none
2023-10-18 17:39:46,414 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,414 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-18 17:39:46,414 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,414 ----------------------------------------------------------------------------------------------------
2023-10-18 17:39:46,414 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 17:39:47,622 epoch 1 - iter 89/894 - loss 3.56088883 - time (sec): 1.21 - samples/sec: 6861.97 - lr: 0.000005 - momentum: 0.000000
2023-10-18 17:39:48,995 epoch 1 - iter 178/894 - loss 3.27510972 - time (sec): 2.58 - samples/sec: 6479.87 - lr: 0.000010 - momentum: 0.000000
2023-10-18 17:39:50,373 epoch 1 - iter 267/894 - loss 2.81223450 - time (sec): 3.96 - samples/sec: 6392.20 - lr: 0.000015 - momentum: 0.000000
2023-10-18 17:39:51,750 epoch 1 - iter 356/894 - loss 2.34160427 - time (sec): 5.34 - samples/sec: 6341.25 - lr: 0.000020 - momentum: 0.000000
2023-10-18 17:39:53,122 epoch 1 - iter 445/894 - loss 1.99483895 - time (sec): 6.71 - samples/sec: 6435.14 - lr: 0.000025 - momentum: 0.000000
2023-10-18 17:39:54,518 epoch 1 - iter 534/894 - loss 1.76339110 - time (sec): 8.10 - samples/sec: 6399.97 - lr: 0.000030 - momentum: 0.000000
2023-10-18 17:39:55,900 epoch 1 - iter 623/894 - loss 1.60438052 - time (sec): 9.49 - samples/sec: 6363.56 - lr: 0.000035 - momentum: 0.000000
2023-10-18 17:39:57,316 epoch 1 - iter 712/894 - loss 1.46822053 - time (sec): 10.90 - samples/sec: 6361.48 - lr: 0.000040 - momentum: 0.000000
2023-10-18 17:39:58,713 epoch 1 - iter 801/894 - loss 1.37216087 - time (sec): 12.30 - samples/sec: 6342.23 - lr: 0.000045 - momentum: 0.000000
2023-10-18 17:40:00,082 epoch 1 - iter 890/894 - loss 1.29129924 - time (sec): 13.67 - samples/sec: 6295.80 - lr: 0.000050 - momentum: 0.000000
2023-10-18 17:40:00,144 ----------------------------------------------------------------------------------------------------
2023-10-18 17:40:00,144 EPOCH 1 done: loss 1.2874 - lr: 0.000050
2023-10-18 17:40:02,389 DEV : loss 0.42374274134635925 - f1-score (micro avg) 0.0047
2023-10-18 17:40:02,417 saving best model
2023-10-18 17:40:02,449 ----------------------------------------------------------------------------------------------------
2023-10-18 17:40:03,852 epoch 2 - iter 89/894 - loss 0.53456617 - time (sec): 1.40 - samples/sec: 6761.96 - lr: 0.000049 - momentum: 0.000000
2023-10-18 17:40:05,224 epoch 2 - iter 178/894 - loss 0.48876177 - time (sec): 2.77 - samples/sec: 6667.46 - lr: 0.000049 - momentum: 0.000000
2023-10-18 17:40:06,685 epoch 2 - iter 267/894 - loss 0.48305597 - time (sec): 4.24 - samples/sec: 6520.37 - lr: 0.000048 - momentum: 0.000000
2023-10-18 17:40:08,045 epoch 2 - iter 356/894 - loss 0.47711486 - time (sec): 5.60 - samples/sec: 6263.90 - lr: 0.000048 - momentum: 0.000000
2023-10-18 17:40:09,415 epoch 2 - iter 445/894 - loss 0.46925640 - time (sec): 6.97 - samples/sec: 6328.63 - lr: 0.000047 - momentum: 0.000000
2023-10-18 17:40:10,812 epoch 2 - iter 534/894 - loss 0.45601089 - time (sec): 8.36 - samples/sec: 6251.92 - lr: 0.000047 - momentum: 0.000000
2023-10-18 17:40:12,213 epoch 2 - iter 623/894 - loss 0.45694794 - time (sec): 9.76 - samples/sec: 6209.82 - lr: 0.000046 - momentum: 0.000000
2023-10-18 17:40:13,604 epoch 2 - iter 712/894 - loss 0.45374120 - time (sec): 11.15 - samples/sec: 6209.76 - lr: 0.000046 - momentum: 0.000000
2023-10-18 17:40:14,984 epoch 2 - iter 801/894 - loss 0.45183281 - time (sec): 12.53 - samples/sec: 6213.64 - lr: 0.000045 - momentum: 0.000000
2023-10-18 17:40:16,315 epoch 2 - iter 890/894 - loss 0.44611403 - time (sec): 13.87 - samples/sec: 6220.24 - lr: 0.000044 - momentum: 0.000000
2023-10-18 17:40:16,371 ----------------------------------------------------------------------------------------------------
2023-10-18 17:40:16,371 EPOCH 2 done: loss 0.4465 - lr: 0.000044
2023-10-18 17:40:21,629 DEV : loss 0.33446836471557617 - f1-score (micro avg) 0.2825
2023-10-18 17:40:21,655 saving best model
2023-10-18 17:40:21,692 ----------------------------------------------------------------------------------------------------
2023-10-18 17:40:23,074 epoch 3 - iter 89/894 - loss 0.37382829 - time (sec): 1.38 - samples/sec: 5689.22 - lr: 0.000044 - momentum: 0.000000
2023-10-18 17:40:24,471 epoch 3 - iter 178/894 - loss 0.38391719 - time (sec): 2.78 - samples/sec: 6001.80 - lr: 0.000043 - momentum: 0.000000
2023-10-18 17:40:25,842 epoch 3 - iter 267/894 - loss 0.38351814 - time (sec): 4.15 - samples/sec: 5934.83 - lr: 0.000043 - momentum: 0.000000
2023-10-18 17:40:27,227 epoch 3 - iter 356/894 - loss 0.37200370 - time (sec): 5.53 - samples/sec: 6050.91 - lr: 0.000042 - momentum: 0.000000
2023-10-18 17:40:28,617 epoch 3 - iter 445/894 - loss 0.36806798 - time (sec): 6.92 - samples/sec: 6106.77 - lr: 0.000042 - momentum: 0.000000
2023-10-18 17:40:30,026 epoch 3 - iter 534/894 - loss 0.36726729 - time (sec): 8.33 - samples/sec: 6160.71 - lr: 0.000041 - momentum: 0.000000
2023-10-18 17:40:31,399 epoch 3 - iter 623/894 - loss 0.36063396 - time (sec): 9.71 - samples/sec: 6126.60 - lr: 0.000041 - momentum: 0.000000
2023-10-18 17:40:32,762 epoch 3 - iter 712/894 - loss 0.36575094 - time (sec): 11.07 - samples/sec: 6135.09 - lr: 0.000040 - momentum: 0.000000
2023-10-18 17:40:34,215 epoch 3 - iter 801/894 - loss 0.36442893 - time (sec): 12.52 - samples/sec: 6201.81 - lr: 0.000039 - momentum: 0.000000
2023-10-18 17:40:35,641 epoch 3 - iter 890/894 - loss 0.36483048 - time (sec): 13.95 - samples/sec: 6178.05 - lr: 0.000039 - momentum: 0.000000
2023-10-18 17:40:35,698 ----------------------------------------------------------------------------------------------------
2023-10-18 17:40:35,698 EPOCH 3 done: loss 0.3646 - lr: 0.000039
2023-10-18 17:40:40,927 DEV : loss 0.3075636327266693 - f1-score (micro avg) 0.3335
2023-10-18 17:40:40,953 saving best model
2023-10-18 17:40:40,988 ----------------------------------------------------------------------------------------------------
2023-10-18 17:40:42,396 epoch 4 - iter 89/894 - loss 0.33269146 - time (sec): 1.41 - samples/sec: 6501.63 - lr: 0.000038 - momentum: 0.000000
2023-10-18 17:40:43,794 epoch 4 - iter 178/894 - loss 0.34094448 - time (sec): 2.81 - samples/sec: 6329.83 - lr: 0.000038 - momentum: 0.000000
2023-10-18 17:40:45,184 epoch 4 - iter 267/894 - loss 0.34400084 - time (sec): 4.20 - samples/sec: 6379.06 - lr: 0.000037 - momentum: 0.000000
2023-10-18 17:40:46,582 epoch 4 - iter 356/894 - loss 0.33882719 - time (sec): 5.59 - samples/sec: 6391.37 - lr: 0.000037 - momentum: 0.000000
2023-10-18 17:40:47,931 epoch 4 - iter 445/894 - loss 0.33813656 - time (sec): 6.94 - samples/sec: 6345.08 - lr: 0.000036 - momentum: 0.000000
2023-10-18 17:40:49,322 epoch 4 - iter 534/894 - loss 0.33235534 - time (sec): 8.33 - samples/sec: 6301.14 - lr: 0.000036 - momentum: 0.000000
2023-10-18 17:40:50,663 epoch 4 - iter 623/894 - loss 0.32836804 - time (sec): 9.67 - samples/sec: 6326.35 - lr: 0.000035 - momentum: 0.000000
2023-10-18 17:40:51,909 epoch 4 - iter 712/894 - loss 0.32290979 - time (sec): 10.92 - samples/sec: 6372.20 - lr: 0.000034 - momentum: 0.000000
2023-10-18 17:40:53,313 epoch 4 - iter 801/894 - loss 0.32823727 - time (sec): 12.32 - samples/sec: 6320.06 - lr: 0.000034 - momentum: 0.000000
2023-10-18 17:40:54,663 epoch 4 - iter 890/894 - loss 0.32666126 - time (sec): 13.67 - samples/sec: 6305.84 - lr: 0.000033 - momentum: 0.000000
2023-10-18 17:40:54,722 ----------------------------------------------------------------------------------------------------
2023-10-18 17:40:54,722 EPOCH 4 done: loss 0.3266 - lr: 0.000033
2023-10-18 17:41:00,006 DEV : loss 0.3056620657444 - f1-score (micro avg) 0.344
2023-10-18 17:41:00,032 saving best model
2023-10-18 17:41:00,066 ----------------------------------------------------------------------------------------------------
2023-10-18 17:41:01,447 epoch 5 - iter 89/894 - loss 0.28380145 - time (sec): 1.38 - samples/sec: 5916.31 - lr: 0.000033 - momentum: 0.000000
2023-10-18 17:41:02,817 epoch 5 - iter 178/894 - loss 0.30119921 - time (sec): 2.75 - samples/sec: 5831.65 - lr: 0.000032 - momentum: 0.000000
2023-10-18 17:41:04,183 epoch 5 - iter 267/894 - loss 0.28339258 - time (sec): 4.12 - samples/sec: 5840.96 - lr: 0.000032 - momentum: 0.000000
2023-10-18 17:41:05,570 epoch 5 - iter 356/894 - loss 0.28951457 - time (sec): 5.50 - samples/sec: 6113.84 - lr: 0.000031 - momentum: 0.000000
2023-10-18 17:41:06,973 epoch 5 - iter 445/894 - loss 0.28772870 - time (sec): 6.91 - samples/sec: 6168.89 - lr: 0.000031 - momentum: 0.000000
2023-10-18 17:41:08,411 epoch 5 - iter 534/894 - loss 0.28686799 - time (sec): 8.34 - samples/sec: 6214.59 - lr: 0.000030 - momentum: 0.000000
2023-10-18 17:41:09,777 epoch 5 - iter 623/894 - loss 0.28834965 - time (sec): 9.71 - samples/sec: 6237.61 - lr: 0.000029 - momentum: 0.000000
2023-10-18 17:41:11,142 epoch 5 - iter 712/894 - loss 0.29652767 - time (sec): 11.08 - samples/sec: 6243.32 - lr: 0.000029 - momentum: 0.000000
2023-10-18 17:41:12,534 epoch 5 - iter 801/894 - loss 0.29303123 - time (sec): 12.47 - samples/sec: 6200.43 - lr: 0.000028 - momentum: 0.000000
2023-10-18 17:41:13,939 epoch 5 - iter 890/894 - loss 0.29497246 - time (sec): 13.87 - samples/sec: 6214.23 - lr: 0.000028 - momentum: 0.000000
2023-10-18 17:41:13,998 ----------------------------------------------------------------------------------------------------
2023-10-18 17:41:13,998 EPOCH 5 done: loss 0.2971 - lr: 0.000028
2023-10-18 17:41:18,978 DEV : loss 0.3028002083301544 - f1-score (micro avg) 0.3519
2023-10-18 17:41:19,003 saving best model
2023-10-18 17:41:19,038 ----------------------------------------------------------------------------------------------------
2023-10-18 17:41:20,430 epoch 6 - iter 89/894 - loss 0.30126272 - time (sec): 1.39 - samples/sec: 6024.34 - lr: 0.000027 - momentum: 0.000000
2023-10-18 17:41:21,869 epoch 6 - iter 178/894 - loss 0.27254140 - time (sec): 2.83 - samples/sec: 6564.55 - lr: 0.000027 - momentum: 0.000000
2023-10-18 17:41:23,268 epoch 6 - iter 267/894 - loss 0.25649955 - time (sec): 4.23 - samples/sec: 6363.86 - lr: 0.000026 - momentum: 0.000000
2023-10-18 17:41:24,660 epoch 6 - iter 356/894 - loss 0.26560972 - time (sec): 5.62 - samples/sec: 6220.58 - lr: 0.000026 - momentum: 0.000000
2023-10-18 17:41:26,014 epoch 6 - iter 445/894 - loss 0.27644281 - time (sec): 6.98 - samples/sec: 6234.12 - lr: 0.000025 - momentum: 0.000000
2023-10-18 17:41:27,384 epoch 6 - iter 534/894 - loss 0.27683306 - time (sec): 8.35 - samples/sec: 6205.24 - lr: 0.000024 - momentum: 0.000000
2023-10-18 17:41:28,745 epoch 6 - iter 623/894 - loss 0.27738781 - time (sec): 9.71 - samples/sec: 6201.08 - lr: 0.000024 - momentum: 0.000000
2023-10-18 17:41:30,134 epoch 6 - iter 712/894 - loss 0.27549257 - time (sec): 11.10 - samples/sec: 6217.29 - lr: 0.000023 - momentum: 0.000000
2023-10-18 17:41:31,587 epoch 6 - iter 801/894 - loss 0.27684533 - time (sec): 12.55 - samples/sec: 6196.98 - lr: 0.000023 - momentum: 0.000000
2023-10-18 17:41:32,991 epoch 6 - iter 890/894 - loss 0.27758019 - time (sec): 13.95 - samples/sec: 6178.48 - lr: 0.000022 - momentum: 0.000000
2023-10-18 17:41:33,050 ----------------------------------------------------------------------------------------------------
2023-10-18 17:41:33,050 EPOCH 6 done: loss 0.2773 - lr: 0.000022
2023-10-18 17:41:38,288 DEV : loss 0.30895572900772095 - f1-score (micro avg) 0.3675
2023-10-18 17:41:38,314 saving best model
2023-10-18 17:41:38,348 ----------------------------------------------------------------------------------------------------
2023-10-18 17:41:39,740 epoch 7 - iter 89/894 - loss 0.23945029 - time (sec): 1.39 - samples/sec: 6134.96 - lr: 0.000022 - momentum: 0.000000
2023-10-18 17:41:41,103 epoch 7 - iter 178/894 - loss 0.24472136 - time (sec): 2.75 - samples/sec: 6145.87 - lr: 0.000021 - momentum: 0.000000
2023-10-18 17:41:42,459 epoch 7 - iter 267/894 - loss 0.24899507 - time (sec): 4.11 - samples/sec: 6014.48 - lr: 0.000021 - momentum: 0.000000
2023-10-18 17:41:43,845 epoch 7 - iter 356/894 - loss 0.26290285 - time (sec): 5.50 - samples/sec: 6101.12 - lr: 0.000020 - momentum: 0.000000
2023-10-18 17:41:45,225 epoch 7 - iter 445/894 - loss 0.25558011 - time (sec): 6.88 - samples/sec: 6150.11 - lr: 0.000019 - momentum: 0.000000
2023-10-18 17:41:46,604 epoch 7 - iter 534/894 - loss 0.25236895 - time (sec): 8.26 - samples/sec: 6113.46 - lr: 0.000019 - momentum: 0.000000
2023-10-18 17:41:47,974 epoch 7 - iter 623/894 - loss 0.25861698 - time (sec): 9.63 - samples/sec: 6128.98 - lr: 0.000018 - momentum: 0.000000
2023-10-18 17:41:49,409 epoch 7 - iter 712/894 - loss 0.25853236 - time (sec): 11.06 - samples/sec: 6213.92 - lr: 0.000018 - momentum: 0.000000
2023-10-18 17:41:50,809 epoch 7 - iter 801/894 - loss 0.25843522 - time (sec): 12.46 - samples/sec: 6173.81 - lr: 0.000017 - momentum: 0.000000
2023-10-18 17:41:52,190 epoch 7 - iter 890/894 - loss 0.26022956 - time (sec): 13.84 - samples/sec: 6227.19 - lr: 0.000017 - momentum: 0.000000
2023-10-18 17:41:52,253 ----------------------------------------------------------------------------------------------------
2023-10-18 17:41:52,254 EPOCH 7 done: loss 0.2598 - lr: 0.000017
2023-10-18 17:41:57,547 DEV : loss 0.30144089460372925 - f1-score (micro avg) 0.3794
2023-10-18 17:41:57,573 saving best model
2023-10-18 17:41:57,606 ----------------------------------------------------------------------------------------------------
2023-10-18 17:41:58,998 epoch 8 - iter 89/894 - loss 0.25238249 - time (sec): 1.39 - samples/sec: 5901.66 - lr: 0.000016 - momentum: 0.000000
2023-10-18 17:42:00,380 epoch 8 - iter 178/894 - loss 0.24111625 - time (sec): 2.77 - samples/sec: 6095.35 - lr: 0.000016 - momentum: 0.000000
2023-10-18 17:42:01,751 epoch 8 - iter 267/894 - loss 0.24040542 - time (sec): 4.14 - samples/sec: 5994.20 - lr: 0.000015 - momentum: 0.000000
2023-10-18 17:42:03,200 epoch 8 - iter 356/894 - loss 0.24702978 - time (sec): 5.59 - samples/sec: 5934.25 - lr: 0.000014 - momentum: 0.000000
2023-10-18 17:42:04,569 epoch 8 - iter 445/894 - loss 0.24595873 - time (sec): 6.96 - samples/sec: 6008.26 - lr: 0.000014 - momentum: 0.000000
2023-10-18 17:42:05,978 epoch 8 - iter 534/894 - loss 0.25606875 - time (sec): 8.37 - samples/sec: 6045.47 - lr: 0.000013 - momentum: 0.000000
2023-10-18 17:42:07,402 epoch 8 - iter 623/894 - loss 0.24760570 - time (sec): 9.79 - samples/sec: 6192.33 - lr: 0.000013 - momentum: 0.000000
2023-10-18 17:42:08,807 epoch 8 - iter 712/894 - loss 0.25033713 - time (sec): 11.20 - samples/sec: 6203.69 - lr: 0.000012 - momentum: 0.000000
2023-10-18 17:42:10,172 epoch 8 - iter 801/894 - loss 0.25087722 - time (sec): 12.57 - samples/sec: 6206.95 - lr: 0.000012 - momentum: 0.000000
2023-10-18 17:42:11,599 epoch 8 - iter 890/894 - loss 0.24848771 - time (sec): 13.99 - samples/sec: 6164.42 - lr: 0.000011 - momentum: 0.000000
2023-10-18 17:42:11,676 ----------------------------------------------------------------------------------------------------
2023-10-18 17:42:11,676 EPOCH 8 done: loss 0.2483 - lr: 0.000011
2023-10-18 17:42:16,927 DEV : loss 0.29879170656204224 - f1-score (micro avg) 0.3868
2023-10-18 17:42:16,953 saving best model
2023-10-18 17:42:16,987 ----------------------------------------------------------------------------------------------------
2023-10-18 17:42:18,387 epoch 9 - iter 89/894 - loss 0.23595464 - time (sec): 1.40 - samples/sec: 5922.44 - lr: 0.000011 - momentum: 0.000000
2023-10-18 17:42:19,796 epoch 9 - iter 178/894 - loss 0.23156957 - time (sec): 2.81 - samples/sec: 6073.83 - lr: 0.000010 - momentum: 0.000000
2023-10-18 17:42:21,181 epoch 9 - iter 267/894 - loss 0.24432659 - time (sec): 4.19 - samples/sec: 6056.05 - lr: 0.000009 - momentum: 0.000000
2023-10-18 17:42:22,530 epoch 9 - iter 356/894 - loss 0.23683646 - time (sec): 5.54 - samples/sec: 6078.25 - lr: 0.000009 - momentum: 0.000000
2023-10-18 17:42:23,903 epoch 9 - iter 445/894 - loss 0.23601101 - time (sec): 6.91 - samples/sec: 6083.59 - lr: 0.000008 - momentum: 0.000000
2023-10-18 17:42:25,297 epoch 9 - iter 534/894 - loss 0.23146609 - time (sec): 8.31 - samples/sec: 6160.96 - lr: 0.000008 - momentum: 0.000000
2023-10-18 17:42:26,717 epoch 9 - iter 623/894 - loss 0.23681248 - time (sec): 9.73 - samples/sec: 6207.57 - lr: 0.000007 - momentum: 0.000000
2023-10-18 17:42:28,054 epoch 9 - iter 712/894 - loss 0.23875417 - time (sec): 11.07 - samples/sec: 6173.32 - lr: 0.000007 - momentum: 0.000000
2023-10-18 17:42:29,433 epoch 9 - iter 801/894 - loss 0.23968107 - time (sec): 12.45 - samples/sec: 6178.51 - lr: 0.000006 - momentum: 0.000000
2023-10-18 17:42:30,871 epoch 9 - iter 890/894 - loss 0.24025730 - time (sec): 13.88 - samples/sec: 6210.19 - lr: 0.000006 - momentum: 0.000000
2023-10-18 17:42:30,929 ----------------------------------------------------------------------------------------------------
2023-10-18 17:42:30,930 EPOCH 9 done: loss 0.2399 - lr: 0.000006
2023-10-18 17:42:35,884 DEV : loss 0.30229300260543823 - f1-score (micro avg) 0.3947
2023-10-18 17:42:35,911 saving best model
2023-10-18 17:42:35,946 ----------------------------------------------------------------------------------------------------
2023-10-18 17:42:37,286 epoch 10 - iter 89/894 - loss 0.23297878 - time (sec): 1.34 - samples/sec: 5957.31 - lr: 0.000005 - momentum: 0.000000
2023-10-18 17:42:38,671 epoch 10 - iter 178/894 - loss 0.21576070 - time (sec): 2.72 - samples/sec: 6349.89 - lr: 0.000004 - momentum: 0.000000
2023-10-18 17:42:40,035 epoch 10 - iter 267/894 - loss 0.22503277 - time (sec): 4.09 - samples/sec: 6305.28 - lr: 0.000004 - momentum: 0.000000
2023-10-18 17:42:41,408 epoch 10 - iter 356/894 - loss 0.23222666 - time (sec): 5.46 - samples/sec: 6276.33 - lr: 0.000003 - momentum: 0.000000
2023-10-18 17:42:42,826 epoch 10 - iter 445/894 - loss 0.23341935 - time (sec): 6.88 - samples/sec: 6380.28 - lr: 0.000003 - momentum: 0.000000
2023-10-18 17:42:44,561 epoch 10 - iter 534/894 - loss 0.23166851 - time (sec): 8.61 - samples/sec: 6073.29 - lr: 0.000002 - momentum: 0.000000
2023-10-18 17:42:45,993 epoch 10 - iter 623/894 - loss 0.23662064 - time (sec): 10.05 - samples/sec: 6036.84 - lr: 0.000002 - momentum: 0.000000
2023-10-18 17:42:47,357 epoch 10 - iter 712/894 - loss 0.23735465 - time (sec): 11.41 - samples/sec: 6051.71 - lr: 0.000001 - momentum: 0.000000
2023-10-18 17:42:48,742 epoch 10 - iter 801/894 - loss 0.23696194 - time (sec): 12.80 - samples/sec: 6066.99 - lr: 0.000001 - momentum: 0.000000
2023-10-18 17:42:50,194 epoch 10 - iter 890/894 - loss 0.23560390 - time (sec): 14.25 - samples/sec: 6056.66 - lr: 0.000000 - momentum: 0.000000
2023-10-18 17:42:50,255 ----------------------------------------------------------------------------------------------------
2023-10-18 17:42:50,255 EPOCH 10 done: loss 0.2359 - lr: 0.000000
2023-10-18 17:42:55,223 DEV : loss 0.3022255003452301 - f1-score (micro avg) 0.393
2023-10-18 17:42:55,281 ----------------------------------------------------------------------------------------------------
2023-10-18 17:42:55,281 Loading model from best epoch ...
2023-10-18 17:42:55,359 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
2023-10-18 17:42:57,662
Results:
- F-score (micro) 0.3817
- F-score (macro) 0.188
- Accuracy 0.2493
By class:
precision recall f1-score support
loc 0.5260 0.5772 0.5504 596
pers 0.2013 0.2793 0.2340 333
org 0.0000 0.0000 0.0000 132
time 0.2143 0.1224 0.1558 49
prod 0.0000 0.0000 0.0000 66
micro avg 0.3869 0.3767 0.3817 1176
macro avg 0.1883 0.1958 0.1880 1176
weighted avg 0.3325 0.3767 0.3517 1176
2023-10-18 17:42:57,662 ----------------------------------------------------------------------------------------------------
|